- Jun 15, 2018
-
-
Joshua Smith authored
More code cleanup. Add logic to check that prefetch address supports Hints. Make sure prefetcher isn't included in any configuration by default
-
- Jun 14, 2018
-
-
Joshua Smith authored
-
- Jun 13, 2018
-
-
Derek Pappas authored
Add DescribedSRAM
-
Gleb Gagarin authored
Dmi write delay fix
-
Joshua Smith authored
-
Joshua Smith authored
- Don't time memory latency for acquires which match paused prefetcher entries. - Consider a stride of 0 as a stride mismatch so that prefetcher entries won't train and send out prefetches for the same cache block.
-
- Jun 12, 2018
-
-
Felix Yan authored
-
Wesley W. Terpstra authored
-
Wesley W. Terpstra authored
Add InModuleBody API
-
Gleb Gagarin authored
Remove unnesessary VPI dependency from SimJTAG
-
- Jun 11, 2018
-
-
Wesley W. Terpstra authored
Parameters are not needed to implement this method. Sometimes you might not have any.
-
Wesley W. Terpstra authored
-
- Jun 09, 2018
-
-
Joshua Smith authored
Some code re-factoring to use more scala-y style with Options. Some code cleanup. Use existing IDPool module to track outstanding source IDs and remove SPFReqArray module
-
Wesley W. Terpstra authored
-
Wesley W. Terpstra authored
This new API is strictly more powerful as you can use it from helper functions. The instatiate() API required you to put your extra glue code in one place and it had to be in the LazyModule.
-
Gleb Gagarin authored
-
- Jun 08, 2018
-
-
Andrew Waterman authored
* Add LookupByHartId key to look up properties without breaking dedup * Use LookupByHartId to avoid directly referencing RocketTilesKey
-
Andrew Waterman authored
* Make lrscCycles a core-specific parameter * Increase lrscCycles to work without branch prediction Since an LR/SC sequence can have taken branches/jumps within the loop, Rockets with no branch prediction can fail to meet the forward-progress guarantee. Increase the timeout to handle the case that all instructions hit in the I$ but up to 14 of them are mispredicted branches. (14 is 16 minus the LR and the SC instructions themselves.)
-
Henry Cook authored
-
- Jun 07, 2018
-
-
Wesley W. Terpstra authored
This bug has flown under the radar because most times we use the same project for the config and the top. However, in the case of shells, you use the top=shell and config=design, which are usually in different projects.
-
Wesley W. Terpstra authored
diplomacy: add BundleBridge for cross-module bundles
-
Wesley W. Terpstra authored
-
- Jun 06, 2018
-
-
Wesley W. Terpstra authored
This can be quite useful for plumbing IO from deep inside the hierarchy to the top of the design.
-
Edmond Cote authored
Additional information found in the following discussion: https://github.com/freechipsproject/rocket-chip/commit/41702a950f2fdc9a247b654869bff6a9389a9bb2#commitcomment-29219295
-
- Jun 05, 2018
-
-
Wesley W. Terpstra authored
This will be necessary to be able to decouple the top 'shell' from the internal 'design'. The way this is going to work is that the 'shell' creates Nodes for it's top-level IO which are then connected to arbitrarily deep points in the design.
-
Andrew Waterman authored
Factor out some generic code in HasRocketTiles
-
Sequencer authored
src/main/scala/rocketchip/Configs.scala should be src/main/scala/subsystem/Configs.scala now
-
Gleb Gagarin authored
-
Jacob Chang authored
-
- Jun 04, 2018
-
-
Wesley W. Terpstra authored
-
- Jun 01, 2018
-
-
Gleb Gagarin authored
-
Andrew Waterman authored
-
Andrew Waterman authored
-
Andrew Waterman authored
-
Andrew Waterman authored
-
Andrew Waterman authored
-
Andrew Waterman authored
-
Andrew Waterman authored
The pipeline design is tightly coupled with that of Rocket, so it isn't general enough to be in BaseTile.
-
Andrew Waterman authored
This seems generic enough.
-
Pentin Alexander Sergeevich authored
NBDcache now creates a RAM named data_array_nbdcache__ext instead of _T_39_ext
-